Hein ten Horn wrote:

The math is perfectly describing what is happening in the
course of time at an arbitrary location in the air or in the
medium inside the cochlea. Concerning the varying
amplitude it does a good job.
But does someone (here) actually know how our hearing
system interprets both indistinguishable(!) frequencies (or
even a within a small range rapidly varying frequency) and
how the resulting 'signal' is translated into what we call the
perception?

Evidently the math given above doesn't
reckon with any hearing mechanism at all. Hence it cannot
rule out perceiving an average frequency.

The mathematics doesn't provide the possibility except, as I have
noted, for brief instants of time. There exists no "wave of average
frequency" in the frequency spectrum of the sum of two waves. A
Fourier analysis of the function doesn't reveal one. The ear doesn't
"produce" one. And I can tell you from personal and professional
experience that it does not hear one. (A triad chord would be truly
awful to experience if it did.)

For the rest I don't get your point on a varying period.
From a mathematical point of view the function

sin( pi * (f_2 + f_1) * t )

has a constant frequency of (f_2 + f_1)/2
and a constant period of 2/(f_2 + f_1).
This frequency is indeed the arithmetical average and
it is not affected by a multiplication of the function by
a relatively slow varying amplitude.

Yes. But when multiplied by a sinusoidal function of a different
frequency (as is the actual equation), the amplitude is affected in a
way which varies in both magnitude and sign with time, and which
affects both the peak spacing and the zero crossings differently from
one cycle to the next as a function of relative phase. If one defines
the period of a waveform as the length of one cycle of a waveform,
then this length of time varies in the way I have previously
described. Please consider using Mathematica or your favorite
plotting program to examine this for yourself.

Apart from the mathematical support, I saw the average
frequency mentioned in several books on physics, unfortunately
without further enclosed proof (as far as I remember).

Apart from the mathematical support, that is also what I have found.
However, I believe this usage has been disappearing in recent years as
re-evaluation replaces reiteration as a means for producing text
books. All I can say is that it appears the claim may have been made
by someone without sufficient experience in the particular field. I
can find no support, anecdotal, phenomenological, psychoacoustical, or
mathematical for the contention (repeated by rote from what I can
tell) that the ear hears the average when the two frequencies are
arbitrarily 'close'.

I've never heard it, and I've been playing musical instruments for 47
years, doing audio electronics for almost 30, and physics for the last
20. The notion appears to me to be speculation based upon little more
than a perfunctory analysis of the underlying mathematics.

It might be more reasonable to claim that what is heard is a slight,
slow warble in frequency, back and forth, from one pitch to the other
accompanyied by a corresponding change in volume. But when the beat
frequency is low, the two pitches are so close together that the
difference between them is not discernable.

However, getting some empirical evidence should be a
rather easy piece of work.

Easier to say than do, certainly, but an interesting and enjoyable
endeavor nevertheless. :-)

jk